Today for review I've got the infamous TRUE CPU cooler, if you don't know what the TRUE is then you aren't paying attention.
The Thermalright Ultra 120 eXtreme CPU cooler is very popular among enthusiast circles as the reigning champion of CPU coolers and most people will quickly recommend a TRUE to someone asking.
So today I've got the TRUE for review, and I've put it to the test using both C2D and C2Q processors, is it all that? Well you'll have to read on to find out…
Thermalright Ultra 120 eXtreme CPU Cooler Review
Author: Kristofer Brozio
Sponsor: Thermalright
Tech Specs,Features or the Basic Info:
Thermalright Ultra 120 eXtreme CPU Cooler
Features:
* Quiet and powerful cooling due to multiple heat pipes and large aluminum fin area
* Proprietary bent winglet design to minimize airflow resistance
* Heat pipes soldered to base (nickel plated)and fins for optimum heat transfer
* Include both bolt-thru-board retention brackets for Intel and AMD
Technical Specs:
* Dimension : L63.44 x W132 x H160.5 mm (heatsink only)
* Weight :790g (Heatsink Only)
* Recommended Fan :All 120mm Fan
Processors Compatibility:
Socket 775 Quad Core
Socket AM2+
Socket AM2
Socket 939 (w/Retention kit)
Motherboard compatibility List
(Both Ultra-120A and Ultra-120 eXtreme do not include a backplate for AM2&AM2+ systems. If you have motherboard manufactured by either Asrock or Gigabyte that utilizes the push pin method, you will need to make a separate purchase for an AM2 metal backplate.)
A Better Look at Things
:
As with most Thermalright products the Ultra 120 eXtreme comes in a plain brown box with the Thermalright logo in the center of the top, the main difference here of course is the 'eXtreme' logo off to the right. Thermlaright was also kind enough to include a second set of fan mounts/clips for me to test using dual fans on the Ultra 120 eXtreme.
Inside we find everything nicely protected with styrofoam and a parts box off to the side.
Once we get everything unpacked we find the instruction manual, Thermalright sticker, motherboard mounting hardware and a separate bag of more parts including vibration dampening strips, thermal compound and the spring loaded mounting screws.
The TRUE is a large heatsink to be sure, but as with any Thermalright product it is very well made and looks rather cool.
On the top, between the heatpipes is the Thermalright logo so if you have a window in your case everyone can see you've got a quality heatsink in there and that you're serious about your system cooling.
One thing interesting I found about the TRUE is the angled fins, something I've never seen before on any CPU cooler of this style. It's called a 'bent winglet design' and it's made that way to minimize airflow resistance making the TRUE more effective at cooling your CPU.
The Thermalright Ultra 120 eXtreme has a cooper base but on first glance it does not look that way, that's because it is nickel plated.
The base is not quite a well finished as I would have expected from a Thermalright product, but this is the case with the majority of TRUE CPU Coolers. I've read numerous posts on the forums about how the base of the TRUE needs to be lapped, and these same users reported significant increases in performance of the TRUE once the base was lapped. I've seen many picture of the base after lapping where all of the nickel plating is gone and just a mirror finished copper base is showing.
The base of the TRUE I received appears to be flat but not very smooth, there are deep machining lines that can easily be seen with the naked eye and felt just by rubbing a finger across it. Using a fingernail you can audibly hear a tick-tick-tick and really feel how deep the grooves are.
Installation, Testing and Comparison
Before we get into the testing/installation I just want to note that I did not use the included thermal compound provided by Thermalright I used Arctic Cooling MX-2 as that is what I used for all the other coolers in the comparison charts.
I also did not lap the base of the CPU cooler, it is exactly as I received it. I know many of you out there report significant increases in performance when lapping the base of the TRUE, but I wanted basically an out-of-box experience or comparison.
Yes I have intentions of lapping the TRUE after all this is done and I get time but for now it's stock as it were..
Normally to install the TRUE you'd have to remove the motherboard from your case, but I don't have to as the case I'm using is a Thermaltake Matrix VX all aluminum case; It was very easy for me to cut out a nice square from the motherboard tray to make it so I don't have to remove the mobo from the case. The average person will have to do the standard long tail installation…heh heh
You'll need to have the 'X' bracket, backplate and four spring loaded screws for installation on the LGA775 platform. C2D and C2Q platforms are exactly the same, I just switched out the CPUs to make it comparable. Obviously my installation was fairly easy and quick, when you remove the mobo you'll be adding more steps and more time to the installation but the basic TRUE install is still the same and easy.
I did run across a compatibility issue with the heatsink on the northbridge of my motherboard, it's large and does not allow me to install the TRUE facing front to back, so I had to turn the cooler ninety degrees to complete the installation, this is a common thing though, on many motherboard this will happen.
Before we jump into the testing here are a few pictures of the TRUE installed with different fans:
Installation, Testing and Comparison Continued:
My system consists of:
Gigabyte GA-G33M-DS2R LGA775 Intel G33 Bear Lake mATX Motherboard
2gigs of Crucial Ballistix Tracer PC2-8500 Ram
Connect3d X1800GTO Video Card
1x Hard drive
1x DVD/RW Drive
all housed inside of a Thermaltake Matrix VX PC Case with two 120mm cooling fans (1 intake/1 exhaust)
I used CoreTemp v.96 to get the temperature readings, I enabled the logging feature and averaged out the temperatures to get the results that you will see in the charts.
To achieve load on the CPU I used Orthos Stress Prime, I basically turned it on and let it run for 3 hours at a time, this was done over several days to insure accurate results.
For the Quad Core testing I ran two instances of Orthos with affinity set for two cores each instance.
To get the idle temps I turned the system on and let it sit running nothing but the Windows XP processes for an hour, this was done over several days as well to insure accurate results.
Ambient room temperature during all testing was 19C (+/- .3C)
Since the Ultra 120 eXtreme does not come with a fan so I figured I'd try it with a few different ones I have on hand to see if there's any difference in performance, also included is a graph with coolers that can be used with no fans.
First up for testing is my C2D 6420 CPU.
Testing using no fan:
We can see the Ultra 120 eXtreme is a very capable passive cooling solution.
Next up is is using the TRUE with assorted 120mm fans, these are all connected to the Molex and running at high speed except for the Antec Tri-Cools which were tested at High and Low speeds, Medium speed was no difference over Low speed so the result was omitted. You'll also see that I included results using two Antec Tri-Cool fans as Thermalright was kind enough to include the second set of mounts for testing.
Here are the Specs for the fans used:
Model: IXP-74-14
1400 RPM
Model: SY1225SL12M
1200 rpm
Antec Tri-Cool 120mm fan
3-speeds (only low and high used for testing)
low- 1200rpm
high- 2200 rpm
From the results you can see the TRUE is a great performing cooler overall. Oddly though using two fans actually netted higher temps, this is most likely the result of the fans themselves not being very efficient at moving air.
Next up we have Quad Core testing with my Q6660 CPU, here we are going to see a different story regarding performance…
First up again is no fans:
Here you can see the temps are below the CPU limit but still fairly high for me, others might not care too much as it's still well within limits.
The final test is with fans, here though I've added the Noctua NF-P12 1300RPM fan to the mix.
Here we see close temps for most of the coolers with the Noctua NH-U12P, Thermalright Ultima 90 and the OCZ Vendetta actually having lower temps. Overall though not bad for all concerned.
You might be wondering how the Vendetta pulled off the low temps, it's the fan, it sounds like a small airplane is inside of my case… it pushes a whole lot of air, if you can deal with the noise…
Another observation I have is that the fan mounting system leaves a lot to be desired, there has to be an easier way to mount fans to the TRUE. When you decide to add two fans then it gets even more frustrating to mount them with the fan clips.
As I mentioned earlier many people have reported very good results by lapping their TRUE coolers, and I believe this is the case here as well, the one I received needs to be lapped to improve its performance.
Summary:
The Thermalright Ultra 120 eXtreme is a great choice for a CPU cooler and performs as expected, very well overall.
I'll be honest though it's not as great as I expected it to be, from all of the hype I've read about the TRUE I expected super low temperatures and the TRUE to blow every cooler I've got away, but that didn't happen here in my testing.
As mentioned though I believe there is more performance to be had from this cooler if the base is lapped to improve the surface, so eventually I'll take the time to lap the base of the TRUE and see what results I can get then.
But, I have to be objective here and can't think of what could be or what can happen if the TRUE was lapped, I have to go by the results I get so…
DragonSteelMods gives the Thermalright Ultra 120 eXtreme a 4.5 out of 5 score and our Recommended Award as well.
Pros:
-Cools well
-Well made
-Fairly easy installation
-Looks great
-Can be used with no fan
-Can be used with low airflow fans and achieve good results
Cons:
-Compatibility issues with certain motherboards
-Base needs lapping
-Large and heavy
-Expensive
I would like to thank Thermalright for the chance to review the Ultra 120 eXtreme and for their continued support of DSM.
Discuss, Questions HERE
review#402